Data analysis is ubiquitous. Some data, however, is not numerical and, even if numerical, may be non-linear. Examples of non-numerical data include scanned documents and photographs. The types of analysis that might be useful on such non-numerical data may include compression, character recognition, etc. Computers, of course, only understand numbers so non-numerical data may be converted to numbers for the computer to understand and further process.